Multicenter comparison study of current methods to measure 25-hydroxyvitamin D in serum

► Ten methods to measure serum 25-hydroxyvitamin D were compared with ID–LC–MS/MS. ► Significant bias exists between ID–LC–MS/MS and many, but not all, other 25(OH)D methods. ► The response among different assays for 25(OH)D2 and 3-epi-25(OH)D3 is variable. ► This variable response impedes the use of uniform cut-off values for vitamin D status. ► Our results indicate the need towards further standardizing 25(OH)D assays. Measurement of serum 25-hydroxyvitamin D [25(OH)D] is generally considered to be a reliable indicator of vitamin D status. The recent increase in diversity of 25(OH)D assays prompted us to evaluate the performance of chromatographic methods (two in-house ID–LC–MS/MS and HPLC (ClinRep, Recipe)), a protein binding method (Cobas-25(OH)D-total, Roche) and immunochemical methods (Liaison and RIA (Diasorin), iSYS (IDS), ADVIA Centaur (Siemens), and Architect i1000 and i2000 (Abbott)). Blood was drawn from randomly selected outpatients (N=60) at one site after informed consent. DEQAS and SRM 972 samples were obtained from the scheme organizer and NIST, respectively. Serum aliquots were prepared, frozen and transported to participating centers. Method comparison was performed according to CLSI-EP9 specifications. With these patient samples, and in comparison with ID–LC–MS/MS, Deming regression parameters slope, intercept and R were found to be within the ranges [0.57–1.07], [−1.7 to 6.9nmol/L] and [0.88–0.98], respectively. 25(OH)D2 in DEQAS and SRM samples was fully recognized by chromatographic methods, but only partially by protein binding and immunochemical methods. Chromatographic methods, and to a lesser extent the protein binding assay, showed cross-reactivity with 3-epi-25(OH)D3. Agreement of 25(OH)D assays to ID–LC–MS/MS in sorting patients into distinct 25(OH)D categories varied between 53% and 88%. Significant bias exists between ID–LC–MS/MS and many, but not all, other 25(OH)D assays. The variable response among different assays for 25(OH)D metabolites impedes the use of uniform cut-off values for defining vitamin D status. Our results indicate the need towards further standardizing assays for 25(OH)D measurement.